llm-foundry
llm-foundry copied to clipboard
mosaicml
LLM training code for Databricks foundation models
"How to specify the specific GPU number, such as GPU:3 or GPU:2, in a multi-GPU card setup?"
**I followed the example in the readme file but encountered an error. How can I solve it?** ------------------------------------------------- (mpt) root@autodl-container-b369119e00-b5dabb5c:~/autodl-tmp/llm-foundry/scripts/train# composer train.py yamls/pretrain/mpt-125m.yaml train_loader.dataset.split=train_small eval_loader.dataset.split=val_small Initializing model... cfg.n_params=1.25e+08 Building train...
KeyError: 'attn_pdrop' with t5-small_dolly_sft.yaml when running inference/convert_composer_to_hf.py
I've setup the t5-small_dolly_sft.yaml file to with `run_name: t5-small-dolly` and ran `composer train.py yamls/finetune/t5-small_dolly_sft.yaml train_loader.dataset.split=train` to generate the check point. from the scripts directory, I run: ``` python inference/convert_composer_to_hf.py \...
Hello, Getting the following circular import error when trying to run the `data/packing.py` script using the command below: `python llmfoundry/data/packing.py --yaml-path /home/justin/code/ai/llm-foundry/scripts/train/yamls/finetune/mpt-7b_dolly_sft.yaml` probably worth looking at `llm-foundry/llmfoundry/data/__init__.py` as this might...
so i have this local db with some info and i want to test if this model can actually speak with the database while i drop non technical commands.for example...
Adds precision to eval. Sets MPT to bf16. For some reason, BF16 + FSDP requires mixed_precision: FULL. It works fine without FSDP. FP16 also works fine and gives basically the...
uses https://github.com/mosaicml/llm-foundry/pull/147 as a springboard to updt torch In interactive instance, I installed torch2 req and everything works fine 125M models was getting good (the same) MFU from the same...
Hi, I am exploring how can I provide custom data and fine-tune the model but before I get to that step I wanted to follow the readme and make sure...
Hi all, Upon the installation step of setup, I am getting an error 'Getting requirements to build wheel' I am using the mosaicml/pytorch:1.13.1_cu117-python3.10-ubuntu20.04 docker image as recommended. It appears that...
This project [depends on pynvml](https://github.com/mosaicml/llm-foundry/blob/93e3290fe2c228a50a67546e6191b71db7171ccb/setup.py#L57), where the nvidia-blessed bindings to nvml are in the [nvidia-ml-py](https://pypi.org/project/nvidia-ml-py/) package instead. This is causing problems for the [gpustat project](https://github.com/wookayin/gpustat/issues/153#issuecomment-1551036105), which needs some internals from...
